CN109492097A - A kind of corporate news data classification of risks method - Google Patents
A kind of corporate news data classification of risks method Download PDFInfo
- Publication number
- CN109492097A CN109492097A CN201811239290.XA CN201811239290A CN109492097A CN 109492097 A CN109492097 A CN 109492097A CN 201811239290 A CN201811239290 A CN 201811239290A CN 109492097 A CN109492097 A CN 109492097A
- Authority
- CN
- China
- Prior art keywords
- news
- sentence
- classification
- data
- enterprise
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 238000000034 method Methods 0.000 title claims abstract description 23
- 239000000463 material Substances 0.000 claims abstract description 29
- 235000012149 noodles Nutrition 0.000 claims abstract description 9
- 239000000284 extract Substances 0.000 claims abstract description 4
- 238000013145 classification model Methods 0.000 claims description 5
- 238000002360 preparation method Methods 0.000 claims description 4
- 238000000605 extraction Methods 0.000 abstract description 3
- 238000005516 engineering process Methods 0.000 description 6
- 230000001419 dependent effect Effects 0.000 description 2
- 230000002349 favourable effect Effects 0.000 description 2
- 238000001914 filtration Methods 0.000 description 2
- 238000000151 deposition Methods 0.000 description 1
- 230000002996 emotional effect Effects 0.000 description 1
- JEIPFZHSYJVQDO-UHFFFAOYSA-N ferric oxide Chemical compound O=[Fe]O[Fe]=O JEIPFZHSYJVQDO-UHFFFAOYSA-N 0.000 description 1
- PCHJSUWPFVWCPO-UHFFFAOYSA-N gold Chemical compound [Au] PCHJSUWPFVWCPO-UHFFFAOYSA-N 0.000 description 1
- 239000010931 gold Substances 0.000 description 1
- 229910052737 gold Inorganic materials 0.000 description 1
- 238000012502 risk assessment Methods 0.000 description 1
- 238000001228 spectrum Methods 0.000 description 1
- 230000029305 taxis Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/20—Natural language analysis
- G06F40/205—Parsing
- G06F40/211—Syntactic parsing, e.g. based on context-free grammar [CFG] or unification grammars
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
Abstract
The invention discloses a kind of corporate news data classification of risks methods, include the following steps: that the Business Name according to determining enterprise obtains the association attributes of the determination enterprise, it scans for the association attributes combination of two and as keyword, news material relevant to the determination enterprise is obtained, and extracts the sentence containing the association attributes from the news material；In sentence inputting CNN sentence disaggregated model containing the association attributes, the sentence classification of each sentence will be obtained, the sentence is classified as positive classification or negative classification；Each sentence classification is weighted processing respectively, takes news category of the big person of sentence classification value as Present News that weight that treated, the news category is positive, and noodles are other or negative classification；The present invention carries out sentence extraction according to enterprise dominant, is predicted by distich subclassification, and then realizes the class prediction for being directed to the news material of the main body.
Description
Technical field
The invention belongs to technical field of data processing, and in particular to a kind of corporate news data classification of risks method.
Background technique
Currently, state-of-the-art technology has a large amount of textual classification model and sentiment analysis model, algorithm is all comparatively mature.
Existing textual classification model and sentiment analysis model are mutually independent algorithm.The mainstream that wherein textual classification model uses is calculated
Method has Bi-LSTM algorithm and CNN, FastText algorithm, can be instructed based on character, word-based being used as entire chapter news
Practice corpus data, since it is used as training corpus for full text, then for only one classification of a specific news article,
But when occurring multiple company's main bodys in news, in fact there may be different points for different company's main bodys
Class.For example, certain news content describes the negative information of company A and the positive information of company B, if divided for full text
Class, can only obtain a classification always, which may be pair for the classification of company A, but company A's and company B
In the different situation of classification (the company A noodles that are negative are other, and the company B noodles that are positive are other), existing classification thinking is unable to satisfy always
Classify in same piece news for different subjects mark.And sentiment analysis relatively mostly uses Bi-LSTM algorithm, sentiment analysis is usual
Only output entire article Sentiment orientation, including front probability, negative probability；There is no more specifical emotional category to distinguish.Therefore,
It is completely dependent on a model prediction, accuracy is highly dependent on the preparation of news corpus data, and it is various in view of news style, together
It is entirely different that the news of sample comes from the possible style of different writer, therefore has limitation.
Summary of the invention
In order to solve the above problems existing in the present technology, it is an object of that present invention to provide one kind can be directed to a certain specific master
The corporate news data classification of risks method that body is classified.
The technical scheme adopted by the invention is as follows:
A kind of corporate news data classification of risks method, includes the following steps:
The association attributes that the determination enterprise is obtained according to the Business Name of determining enterprise, by the association attributes combination of two
And scanned for as keyword, news material relevant to the determination enterprise is obtained, and extract from the news material
Contain the sentence of the association attributes out；
In sentence inputting CNN sentence disaggregated model containing the association attributes, the sentence classification of each sentence will be obtained,
The sentence is classified as positive classification or negative classification；
Each sentence classification is weighted processing respectively, takes the big person of sentence classification value that weights that treated as working as
The news category of preceding news, the news category is positive, and noodles are other or negative classification.
Further, the association attributes include but is not limited to that method name, Gao Guanming, company's abbreviation, stock abbreviation, company are gone through
History name and ProductName.
Further, the CNN sentence disaggregated model is the Company News disaggregated model made of the training of CNN algorithm.
Further, training forms the CNN sentence disaggregated model with the following method:
Prepare training corpus data；
By in the sentence inputting CNN sentence classification based training model in training corpus data, training obtains CNN sentence classification mould
Type.
Further, the preparation training corpus data include the following steps:
Grab enterprise-class news material in news data source using web crawlers, and by the enterprise-class news material with
The form storage of text is in the database；
According to the news focus of enterprises pay attention, news category needed for counting is summarized；
For the different customized a series of strong rules of news category；
According to the customized strong rule, the news material conduct to match with the strong rule is filtered out in the database
Spare corpus data；
Using manually to strong rule sift out come spare corpus data check, filter out the first training corpus data；
Using the data for manually obtaining different news categories from major website, as the second training corpus data；
First corpus data and the second corpus data are merged, training corpus data are obtained.
The invention has the benefit that
The present invention carries out sentence extraction according to enterprise dominant, is predicted by distich subclassification, and then realizes and be directed to
The class prediction of the news material of the main body.Since each sentence includes the association attributes of determining enterprise, prediction result
Necessarily it is directed to the determination enterprise.If multiple enterprise dominants involved in same piece news material can using the method for the present invention
Different sentences is extracted according to different subjects, obtains the news category for being directed to different enterprise dominants, and it is more accurate to classify.
Detailed description of the invention
Fig. 1 is flow chart of the present invention.
Fig. 2 is to prepare training corpus data flowchart.
Specific embodiment
With reference to the accompanying drawing and specific embodiment the present invention is further elaborated.Following embodiment is only used for clearer
Ground illustrates product of the invention, therefore is intended only as example, and not intended to limit the protection scope of the present invention.
Embodiment:
A kind of corporate news data classification of risks method provided in an embodiment of the present invention, as shown in Figure 1, including following step
It is rapid:
S101, the association attributes that the determination enterprise is obtained according to the Business Name of determining enterprise, by the association attributes two
Two groups of merging are scanned for as keyword, obtain relevant to determination enterprise news material, and from the news material
In extract the sentence containing the association attributes.
It determines that enterprise is the enterprise for needing to carry out news data risk analysis, is obtained according to the Business Name of the determination enterprise
The association attributes of the determination enterprise, association attributes include but is not limited to method name, Gao Guanming, company's abbreviation, stock abbreviation, company
History name and ProductName.
The meaning of combination of two is the relationships that two association attributes are and.Using the association attributes of combination of two as keyword
The search of news material is carried out, accuracy is higher, can prevent from searching because of the appearance of different company's same alike result value and be somebody's turn to do
It determines the incoherent news material of enterprise, influences subsequent calculating.For example, Chongqing Yu Cun big data Science and Technology Ltd. and Beijing reputation
The company for depositing big data Science and Technology Ltd. is referred to as possible to deposit big data for reputation, if only carried out with single association attributes
Search, then the news material that can not be accurately positioned in search result is about Chongqing Yu Cun big data Science and Technology Ltd. or north
Jing Yucun big data Science and Technology Ltd..
The association attributes combination of two that will determine enterprise, and scans on the internet as keyword, obtain with
The relevant news material of determination enterprise, and extracted from the news material and contain the determination enterprise association attributes (keyword)
Sentence.
S102, by containing the association attributes sentence inputting CNN sentence disaggregated model in, obtain the sentence of each sentence
Classification, the sentence are classified as positive classification or negative classification.
CNN sentence disaggregated model is the Company News disaggregated model made of the training of CNN algorithm, which can be used existing
There is the training of textual classification model training method to form.Each sentence classification is predicted by CNN sentence disaggregated model, is obtained
The classification of each sentence, this is classified as positive classification or negative classification.Since each sentence contains the association attributes of determining enterprise,
Therefore, the prediction of sentence classification is the prediction carried out for the determination enterprise.
S103, each sentence classification is weighted processing respectively, takes the big person of sentence classification value that weights that treated
As the news category of Present News, the news category is positive, and noodles are other or negative classification.
In the present embodiment, 3 are assigned by headline weight, remaining equal weight assigns 1, because headline is often more
Represent the Sentiment orientation of author.Sentence classification each in news material is added after weighting is handled respectively, the big person of value is used as should
The news category of news material.It will be added after the sentence of positive classification and the sentence of negative classification respectively weighting processing, if just
The other value of noodles is big, then the news category noodles that are positive are other, if the value of negative classification is big, the news category noodles that are negative are other.
The present invention carries out sentence extraction according to enterprise dominant, is predicted by distich subclassification, and then realizes and be directed to
The class prediction of the news material of the main body.Since each sentence includes the association attributes of determining enterprise, prediction result
Necessarily it is directed to the determination enterprise.If multiple enterprise dominants involved in same piece news material can using the method for the present invention
Different sentences is extracted according to different subjects, obtains the news category for being directed to different enterprise dominants, and it is more accurate to classify.
The present invention is predicted only for enterprise-class news (finance and economics plate, the company's plate of such as news), passes through combination
CNN sentence disaggregated model predicts news data risk, and the wind of enterprise dominant in news can be more accurately predicted
Dangerous information, accuracy are higher.
Training CNN sentence disaggregated model be unable to do without training corpus, referring to fig. 2: in the present invention, training corpus data preparation side
Method includes the following steps:
S201, enterprise-class news material as much as possible is grabbed in news data source using web crawlers, and should
Enterprise-class news material stores in the database in a text form.
News data source include the major portal website in the whole nation corporate news and financial and economic news plate and with finance and economics, enterprise
Relevant each middle-size and small-size website such as industry.
S202, the news focus according to enterprises pay attention summarize news category needed for counting.
News category includes but is not limited to " tax evasion ", " Policy Supervision ", " risk of breaking one's promise ", " delinquent ", " accident
Information ", " product problem ", " win-win cooperation ", " business variation ", " plagiarizing infringement ", " disputes act ", " violates " equity variation "
Regulation ", " wage arrears ", " product up-gradation ", " senior executive departing ", " investment and financing ", " operations risks ", " absconding to avoid punishment ", " corruption
Bribe ", " fraud fraud ", " achievement awards ", " cuts in salaries of reducing the staff ", " listing failure ", " stock is favourable ", " break ", " strategy
Risk ", " disclosing wrong ", " bulletin publicity ", " mortgage is pledged ", " stop doing business rectification ", " stock empty profit ", " debt information ", " achievement
Loss ", " financial risk ", " business debt ", " other ", " Cooperation Risk ".
Most news categories are risk classification, such as tax evasion, intuitively embody news and describe mainstream corporation
Negative information, so that user has a basic understanding to main body enterprise.
S203, for the different customized a series of strong rules of news category.
Strong rule is configured according to the actual situation, such as achievement awards, setting rule are as follows: ' praise .* achievement | (year
Degree | Forbes) .* (list | personage | collective | manager) | (obtain | win | authorize | selected) .* (unit " | unit " | "unity enterprise" |
Enterprise " | company " | "tibco software, inc." "TIBCO Software | patent | prize (gold) | title | honor | " degree | doctor | personage | manager | collective) | (annual report |
China | enterprise | the whole world | the world) .* (it is strong < strong | list | name company | best | seniority among brothers and sisters of paying taxes) | (enter | rank among) .* (world | in
State | area) .* it is strong | (human resources | strong) .* ranking list | " continue to hold a post or title | it is best to obtain .* | the .* that is shortlisted for (strong | list) | jump .* is the first | valence
Value list .* publication | enter .* list | global .* maximum .* platform | publication .* unicorn list | sell ground .* first | it is honoured with year | wealth
Rich .* changes the company in the world | ' climb to the top of a mountain more than .* | the net profit .* industry umber one | the property .* that rises suddenly and sharply steps on the richest | and bright spot is prominent |
Matchmaker .* comments .* most beautiful | and success .* is maximum | obtain .* (third place | champion | second place) | keep the steady .* expansion of .* | contest .* bonus | it wins
Obtain .* favorable comment | ' wound .* item first '.
S204, according to the customized strong rule of step S203, filter out in the database match with the strong rule it is new
Material is heard as spare corpus data.
S205, using manually to strong rule sift out come spare corpus data check, filter out the first training corpus
Data.
In a particular embodiment, core manually is carried out to the spare corpus data for specifying strong Rules Filtering to come out as needed
It is right, to determine whether the spare corpus screened belongs to specified news category, prevent strong rule error.Because of news type
Formula is varied, influenced by writer it is quite big, sometimes strong Rules Filtering go out data be all not fully we want
The data taken.The step for increasing artificial nucleus couple, keeps training corpus data more accurate, to guarantee trained model accuracy rate
It is higher.
S206, using the data for manually obtaining different news categories from major website, as the second training corpus data.
S207, the first corpus data and the second corpus data are merged, obtains training corpus data.
In training corpus data, the training corpus data of each news category are no less than 5000.
First training corpus data and the second training corpus data are prepared in 1:1 ratio.And the first training corpus data
It is not repeated with the second training corpus data.
By in the sentence inputting CNN sentence classification based training model in training corpus, using open source CNN algorithm, training is obtained
CNN sentence disaggregated model.
The present invention is not limited to above-mentioned optional embodiment, anyone can show that other are various under the inspiration of the present invention
The product of form, however, make any variation in its shape or structure, it is all to fall into the claims in the present invention confining spectrum
Technical solution, be within the scope of the present invention.
Claims (5)
1. a kind of corporate news data classification of risks method, which comprises the steps of:
The association attributes of the determining enterprise are obtained according to the Business Name of determining enterprise, simultaneously by the association attributes combination of two
It is scanned for as keyword, obtains news material relevant to the determining enterprise, and extract from the news material
Contain the sentence of the association attributes out；
In sentence inputting CNN sentence disaggregated model containing the association attributes, the sentence classification of each sentence will be obtained, it is described
Sentence is classified as positive classification or negative classification；
Each sentence classification is weighted processing respectively, takes weighting treated the big person of sentence classification value as currently new
The news category of news, the news category is positive, and noodles are other or negative classification.
2. corporate news data classification of risks method according to claim 1, which is characterized in that the association attributes include
But it is not limited to method name, Gao Guanming, company's abbreviation, stock abbreviation, corporate history name and ProductName.
3. corporate news data classification of risks method according to claim 1, which is characterized in that the CNN sentence classification
Model is the Company News disaggregated model made of the training of CNN algorithm.
4. corporate news data classification of risks method according to claim 3, which is characterized in that the CNN sentence classification
Training forms model with the following method:
Prepare training corpus data；
By in the sentence inputting CNN sentence classification based training model in training corpus data, training obtains CNN sentence disaggregated model.
5. corporate news data classification of risks method according to claim 4, which is characterized in that the preparation training corpus
Data include the following steps:
Enterprise-class news material is grabbed in news data source using web crawlers, and by the enterprise-class news material with text
This form storage is in the database；
According to the news focus of enterprises pay attention, news category needed for counting is summarized；
For the different customized a series of strong rules of news category；
According to the customized strong rule, the news material to match with the strong rule is filtered out in the database as standby
Use corpus data；
Using manually to strong rule sift out come spare corpus data check, filter out the first training corpus data；
Using the data for manually obtaining different news categories from major website, as the second training corpus data；
First corpus data and the second corpus data are merged, training corpus data are obtained.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811239290.XA CN109492097B (en) | 2018-10-23 | 2018-10-23 | Enterprise news data risk classification method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811239290.XA CN109492097B (en) | 2018-10-23 | 2018-10-23 | Enterprise news data risk classification method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109492097A true CN109492097A (en) | 2019-03-19 |
| CN109492097B CN109492097B (en) | 2021-11-16 |
Family
ID=65692537
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811239290.XA Active CN109492097B (en) | 2018-10-23 | 2018-10-23 | Enterprise news data risk classification method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109492097B (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110298403A (en) * | 2019-07-02 | 2019-10-01 | 郭刚 | The sentiment analysis method and system of enterprise dominant in a kind of financial and economic news |
| CN110502638A (en) * | 2019-08-30 | 2019-11-26 | 重庆誉存大数据科技有限公司 | A kind of Company News classification of risks method based on target entity |
| CN111475646A (en) * | 2020-03-17 | 2020-07-31 | 赵志杰 | Method, device and equipment for evaluating environment image |
| CN111694955A (en) * | 2020-05-08 | 2020-09-22 | 中国科学院计算技术研究所 | Early dispute message detection method and system for social platform |
Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20050138079A1 (en) * | 2003-12-17 | 2005-06-23 | International Business Machines Corporation | Processing, browsing and classifying an electronic document |
| CN102023967A (en) * | 2010-11-11 | 2011-04-20 | 清华大学 | Text emotion classifying method in stock field |
| US20130231920A1 (en) * | 2012-03-02 | 2013-09-05 | Clarabridge, Inc. | Apparatus for identifying root cause using unstructured data |
| CN105205043A (en) * | 2015-08-26 | 2015-12-30 | 苏州大学张家港工业技术研究院 | Classification method and system of emotions of news readers |
| CN106294326A (en) * | 2016-08-23 | 2017-01-04 | 成都科来软件有限公司 | A kind of news report Sentiment orientation analyzes method |
| CN107220237A (en) * | 2017-05-24 | 2017-09-29 | 南京大学 | A kind of method of business entity's Relation extraction based on convolutional neural networks |
| CN107403017A (en) * | 2017-08-09 | 2017-11-28 | 上海数旦信息技术有限公司 | A kind of method that real-time news of intellectual analysis influences on financial market |
| US20180150562A1 (en) * | 2016-11-25 | 2018-05-31 | Cognizant Technology Solutions India Pvt. Ltd. | System and Method for Automatically Extracting and Analyzing Data |
| CN108399230A (en) * | 2018-02-13 | 2018-08-14 | 上海大学 | A kind of Chinese financial and economic news file classification method based on convolutional neural networks |
-
2018
- 2018-10-23 CN CN201811239290.XA patent/CN109492097B/en active Active
Patent Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20050138079A1 (en) * | 2003-12-17 | 2005-06-23 | International Business Machines Corporation | Processing, browsing and classifying an electronic document |
| CN102023967A (en) * | 2010-11-11 | 2011-04-20 | 清华大学 | Text emotion classifying method in stock field |
| US20130231920A1 (en) * | 2012-03-02 | 2013-09-05 | Clarabridge, Inc. | Apparatus for identifying root cause using unstructured data |
| CN105205043A (en) * | 2015-08-26 | 2015-12-30 | 苏州大学张家港工业技术研究院 | Classification method and system of emotions of news readers |
| CN106294326A (en) * | 2016-08-23 | 2017-01-04 | 成都科来软件有限公司 | A kind of news report Sentiment orientation analyzes method |
| US20180150562A1 (en) * | 2016-11-25 | 2018-05-31 | Cognizant Technology Solutions India Pvt. Ltd. | System and Method for Automatically Extracting and Analyzing Data |
| CN107220237A (en) * | 2017-05-24 | 2017-09-29 | 南京大学 | A kind of method of business entity's Relation extraction based on convolutional neural networks |
| CN107403017A (en) * | 2017-08-09 | 2017-11-28 | 上海数旦信息技术有限公司 | A kind of method that real-time news of intellectual analysis influences on financial market |
| CN108399230A (en) * | 2018-02-13 | 2018-08-14 | 上海大学 | A kind of Chinese financial and economic news file classification method based on convolutional neural networks |
Non-Patent Citations (2)
| Title |
|---|
| AYANA YAMAMOTO 等: "Company Relation Extraction from Web News Articles for Analyzing Industry Structure", 《2017 IEEE 11TH INTERNATIONAL CONFERENCE ON SEMANTIC COMPUTING (ICSC)》 * |
| 周邦定: "客户负面新闻自动检索方法的研究与应用", 《中国优秀硕士学位论文全文数据库 信息科技辑》 * |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110298403A (en) * | 2019-07-02 | 2019-10-01 | 郭刚 | The sentiment analysis method and system of enterprise dominant in a kind of financial and economic news |
| CN110298403B (en) * | 2019-07-02 | 2023-12-12 | 北京金融大数据有限公司 | Emotion analysis method and system for enterprise main body in financial news |
| CN110502638A (en) * | 2019-08-30 | 2019-11-26 | 重庆誉存大数据科技有限公司 | A kind of Company News classification of risks method based on target entity |
| CN111475646A (en) * | 2020-03-17 | 2020-07-31 | 赵志杰 | Method, device and equipment for evaluating environment image |
| CN111694955A (en) * | 2020-05-08 | 2020-09-22 | 中国科学院计算技术研究所 | Early dispute message detection method and system for social platform |
| CN111694955B (en) * | 2020-05-08 | 2023-09-12 | 中国科学院计算技术研究所 | Early dispute message detection method and system for social platform |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109492097B (en) | 2021-11-16 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109492097A (en) | A kind of corporate news data classification of risks method | |
| CN109472470A (en) | In conjunction with the corporate news data classification of risks method of deep learning and logic rules | |
| CN110334212A (en) | A kind of territoriality audit knowledge mapping construction method based on machine learning | |
| CN106960063A (en) | A kind of internet information crawl and commending system for field of inviting outside investment | |
| CN109859052A (en) | A kind of intelligent recommendation method, apparatus, storage medium and the server of investment tactics | |
| CN113706185A (en) | Method and system for predicting market behavior based on news and sentiment analysis | |
| CN107704512A (en) | Financial product based on social data recommends method, electronic installation and medium | |
| KR102121901B1 (en) | System for online public fund investment management assessment service | |
| CN107832425A (en) | A kind of corpus labeling method, the apparatus and system of more wheel iteration | |
| CN110309508A (en) | A kind of VWAP quantization transaction system and method based on investor sentiment | |
| CN110502638A (en) | A kind of Company News classification of risks method based on target entity | |
| Duan et al. | Corporate social responsibility information disclosure objective or not: an empirical research of chinese listed companies based on text mining | |
| Bhat et al. | Automated stock price prediction and trading framework for Nifty intraday trading | |
| CN112419029A (en) | Similar financial institution risk monitoring method, risk simulation system and storage medium | |
| CN110096631B (en) | Stock market emotion report generation method based on postings text analysis of stock forum | |
| Peng | Analysis of investor sentiment and stock market volatility trend based on big data strategy | |
| CN108564465A (en) | A kind of enterprise credit management method | |
| Lee et al. | Esg discourse analysis through bertopic: comparing news articles and academic papers | |
| Chen et al. | Analysis of the impact of investor sentiment on stock price using the latent dirichlet allocation topic model | |
| CN109086927A (en) | In conjunction with the multiple-factor method of commerce of big data the analysis of public opinion and Fusion Model | |
| CN110162781A (en) | A kind of finance text subjectivity sentence automatic identifying method | |
| Lee et al. | Crudeoilnews: An annotated crude oil news corpus for event extraction | |
| Chuaykoblap et al. | Expert-based text mining with Delphi method for crude oil price prediction | |
| Lee et al. | An annotated commodity news corpus for event extraction | |
| Klejdysz et al. | Shifts in ECB communication: a text mining approach |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| CP03 | Change of name, title or address |
Address after: 401121 Chongqing Yubei District Huangshan Avenue No. 53 with No. 2 Kirin C Block 9 Floor Patentee after: Chongqing Yucun Technology Co.,Ltd. Country or region after: China Address before: 401121 Chongqing Yubei District Huangshan Avenue No. 53 with No. 2 Kirin C Block 9 Floor Patentee before: CHONGQING SOCIALCREDITS BIG DATA TECHNOLOGY CO.,LTD. Country or region before: China |
|
| CP03 | Change of name, title or address |